Biometrics DOI: 10.1111/j.1541-0420.2007.01039.

Bayesian Distributed Lag Models: Estimating Effects of Particulate

Matter Air Pollution on Daily Mortality

L. J. Welty,1,∗ R. D. Peng,2 S. L. Zeger,2 and F. Dominici2

1
Department of Preventive Medicine, Northwestern University, Feinberg School of Medicine,
680 North Lake Shore Drive, Suite 1102, Chicago, Illinois 60611, U.S.A.
2
Department of Biostatistics, Johns Hopkins Bloomberg School of Public Health,
615 North Wolfe Street, Baltimore, Maryland 21205, U.S.A.
∗
email: lwelty@northwestern.edu

Summary. A distributed lag model (DLagM) is a regression model that includes lagged exposure vari-
ables as covariates; its corresponding distributed lag (DL) function describes the relationship between the
lag and the coefficient of the lagged exposure variable. DLagMs have recently been used in environmental
epidemiology for quantifying the cumulative effects of weather and air pollution on mortality and morbid-
ity. Standard methods for formulating DLagMs include unconstrained, polynomial, and penalized spline
DLagMs. These methods may fail to take full advantage of prior information about the shape of the DL
function for environmental exposures, or for any other exposure with effects that are believed to smoothly
approach zero as lag increases, and are therefore at risk of producing suboptimal estimates. In this article,
we propose a Bayesian DLagM (BDLagM) that incorporates prior knowledge about the shape of the DL
function and also allows the degree of smoothness of the DL function to be estimated from the data. We
apply our BDLagM to its motivating data from the National Morbidity, Mortality, and Air Pollution Study
to estimate the short-term health effects of particulate matter air pollution on mortality from 1987 to 2000
for Chicago, Illinois. In a simulation study, we compare our Bayesian approach with alternative methods
that use unconstrained, polynomial, and penalized spline DLagMs. We also illustrate the connection be-
tween BDLagMs and penalized spline DLagMs. Software for fitting BDLagM models and the data used in
this article are available online.
Key words: Air pollution; Bayes; Distributed lag; Mortality; NMMAPS; Penalized splines; Smoothing;
Time series.

1. Introduction might underestimate the risk of mortality associated with air

Distributed lag models (DLagMs; Almon, 1965) are regression
models that include lagged exposure variables, or distributed
lags (DLs), as covariates. They have recently been employed
in environmental epidemiology for estimating short-term cu-
mulative effects of environmental exposures on daily mortal-
ity or morbidity (e.g., Pope et al., 1991; Pope and Schwartz,
1996; Braga et al., 2001; Zanobetti et al., 2002; Kim, Kim,
and Hong, 2003; Bell McDermott, Zeger, Samet, and Do-
minici, 2004; Goodman, Dockery, and Clancy, 2004; Welty
and Zeger, 2005). DLagMs are specialized types of varying-
coefficient models (Hastie and Tibshirani, 1993) and dynamic
linear models (Ravines, Schmidt, and Migon, 2006).
For Poisson log-linear DLagMs that estimate the effects
of lagged air pollution levels on daily mortality counts, the
sum of the DL coefficients is interpreted as the percentage
increase in daily mortality associated with a one unit in-
crease in air pollution on each of the previous days. Because
the time from exposure to event will almost certainly vary in
a population, this sum is a more appropriate measure of the
effect of short-term exposure than a single day’s coefficient.
Results from previous time series studies suggest that com-
pared to DLagMs, models with single day pollution exposures
pollution (Schwartz, 2000; Zanobetti et al., 2003; Goodman
et al., 2004; Roberts, 2005).
Exposure variables, such as ambient air pollution levels,
may be highly correlated over time, making DL coefficients
difficult to estimate. A general solution is to constrain the co-
efficients as a function of lag. Common constraints include a
polynomial (Almon, 1965) or a spline (Corradi, 1977). Esti-
mating DLagMs as varying-coefficient models constrains the
coefficients to follow a natural cubic spline (Hastie and Tib-
shirani, 1993). The DL function for air pollution and mor-
tality has been estimated with polynomial constraints (e.g.,
Schwartz, 2000, Braga et al., 2001; Kim et al., 2003; Bell,
Samet, and Dominici, 2004; Goodman et al., 2004), spline
constraints (Zanobetti et al., 2000), and without constraints
(Zanobetti et al., 2003).
Each type of constraint on the DL coefficients is an appli-
cation of prior knowledge to model specification. In the con-
text of air pollution and mortality, prior knowledge suggests
that short-term risk of mortality varies smoothly as a func-
tion of lag and decreases to zero. Prior knowledge about the
effects of air pollution on mortality at early lags is limited.
There may be short delays in health effects after exposure,


C 2008, The International Biometric Society 1
2 Biometrics
as suggested by studies of single day pollution exposures that
find the largest effect on mortality at lag day 1 (Zmirou et al.,
1988; Katsouyanni et al., 2001; Dominici et al., 2003). In the
scenario of mortality displacement (Schimmel and Murawsky,
1978), in which high air pollution levels may advance by sev-
eral days the deaths of frail individuals, the DL function may
be zero or positive at early lags, then decrease and become
negative (Zanobetti et al., 2000, 2002). If there were both a
delay in health effect and mortality displacement, hypotheses
concerning the sign or smoothness of the DL function at early
lags would be tenuous at best.
For more appropriate model specification and improved es-
timation, it may be advisable to formulate DLagMs so that
(i) coefficients are constrained to approach zero smoothly
with increasing lag and (ii) early coefficients are relatively
unconstrained. Neither polynomial nor spline constraints, the
most common methods for specifying DLagMs, include this
prior information in estimation. In this article, we develop
Bayesian DLagMs (BDLagMs) that incorporate our under-
standing of the relationship between short-term fluctuations
of particulate matter (PM) air pollution and daily fluctuations
in mortality counts. Our prior distribution specifies that as
lag increases, the DL function will have increasing smooth-
ness and approach zero. An advantage of our approach is
that the degree of smoothness of the DL function is estimated
from the data. We note that BDLagMs have been explored in
economics (e.g., Leamer, 1972; Schiller, 1973; Ravines et al.,
2006), and autoregressive priors have been used generally to
smooth time-dependent coefficients in generalized linear mod-
els (e.g., Fahrmeir and Knorr-Held, 1997; Manda and Meyer,
2005). However, our prior is quite different from those using
a constant degree of smoothness (Schiller, 1973), a particu-
lar parametric form (Leamer, 1972; Ravines et al., 2006), or
an autoregressive structure (e.g., Fahrmeir and Knorr-Held,
1997; Manda and Meyer, 2005).
We apply our BDLagM to data from the National Mor-
bidity, Mortality, and Air Pollution Study (NMMAPS) to es-
timate the shape of the DL function between daily PM and
daily deaths for Chicago, Illinois from 1987 to 2000. We exam-
ine the sensitivity of the estimated DL function to the speci-
fication of the BDLagM prior. We compare the air pollution
effect estimated with the BDLagM to that estimated using
unconstrained maximum likelihood (ML). We also compare
air pollution effects estimated under the full formulation of
the BDLagM, computed using a Gibbs sampler, to those es-
timated under an approximate formulation, computed using
a closed form expression.
We also conduct a simulation study comparing BDLagMs
to unconstrained, polynomial, and penalized spline DLagMs.
For penalized spline DLagMs, we compare estimates obtained
using generalized cross validation (GCV) and restricted maxi-
mum likelihood estimation (REML; Ruppert, Wand, and Car-
roll, 2003). We include DLagMs that are consistent with bi-
ological knowledge along with DLagMs for which our BD-
LagMs may be misspecified.
Because constraining DL coefficients is a way of smooth-
ing, we consider how our Bayesian approach relates to pe-
nalized spline DLagMs. We demonstrate that BDLagMs are
analogous to penalized spline DLagMs with a specific penalty
matrix derived from the BDLagM prior.
Though our BDLagM formulation was motivated by a de-
sire to model flexibly the DL function between lagged PM
levels and daily mortality counts, it is relevant to situations
in which the lagged effects of an exposure on an outcome
are unknown for the first few lags but are believed to dissi-
pate with lag. Using BDLagMs with repeated measures data
would require extensions to our approach. For documenta-
tion and to encourage implementation, our BDLagM soft-
ware is available online at http://www.ihapss.jhsph.edu/
software/BayesDLM/.
2. Bayesian DLagMs
Let y t and x t be the outcome and exposure time series. We
consider a generalized linear DLagM g(E[yt | x1 , . . . , xt ]) =
L
θ x where L is the maximum lag and θ = (θ 0 , . . . , θ L )
=0  t−
is the vector of the DL coefficients to be estimated. Initially
 will consider the normal linear model E[yt | x1 , . . . , xt ] =
we
θ xt− , with Y t independent normal with constant vari-
ance.
The goal is to specify a prior on θ = (θ 0 , θ 1 , . . . , θ L ) that
is uninformative on the DL coefficients for small  but that
constrains the coefficients with larger  to be smoother and ap-
proach zero. We assume θ ∼ N (0, Ω), where Ω is constructed
so that for increasing lag the diagonal elements decrease to
zero (Var(θ  ) → 0) and the off–diagonal elements in its corre-
lation matrix increase to one (Cor(θ −1 , θ  ) → 1). Care must
be taken to construct Ω so that it remains positive definite.
A natural approach is to define Ω = ABA, where AAT is the
diagonal matrix of the individual variances of the θ  s, and B is
the correlation matrix for θ . Specifying an appropriate Ω may
then be achieved by setting A equal to the Cholesky decom-
position of a diagonal matrix with the desired prior variances
and setting B equal to the correlation matrix for increasingly
correlated normal random variables.
To define A, let the parameter σ 2 be the prior variance of
θ 0 , and set Var(θ 1 ) = v 1 σ 2 , . . . , Var(θ L ) = v L σ 2 where the v  s
are a decreasing sequence of weights such that 1 ≥ v 1 ≥ · · · ≥
v L > 0. We parameterize them by v  (η 1 ) = exp(η 1 ), η 1 ≤ 0,
so that the hyperparameter η 1 governs how quickly the prior
variances of the θ  s approach zero. Choosing the exponential
function is convenient but not required. Let V(η 1 ) be the
diagonal matrix with entries 1, v 1 (η 1 )1/2 , . . . , v L (η 1 )1/2 . We
set A = σV(η 1 ).
To specify the correlation matrix B, we similarly define
w  (η 2 ) = exp(η 2 ), η 2 ≤ 0, to be a decreasing sequence of
weights, and M(η 2 ) to be the (L + 1) × (L + 1) diago-
nal matrix with entries 1, w 1 (η 2 ), . . . , w L (η 2 ). We let B =
W(η 2 ), where W(η 2 ) is the correlation matrix derived from
the covariance matrix M(η 2 )M(η 2 ) + {I L+1 − M(η 2 )}1 L+1 ×

1L+1 {I L+1 − M(η 2 )} , where by 1 L+1 we mean a (L + 1) × 1
vector of ones and by I L+1 we mean the (L + 1) × (L + 1)
identity matrix. Then W(η 2 ) is the correlation matrix for
the mixture of normal random variables M(η 2 )X 1 + {I L+1 −
M(η 2 )}1 L+1 X 2 where X 1 ∼ N (0, I L+1 ) and X 2 ∼ N (0, 1).
The first few elements of the independent X 1 are weighted
more heavily than the corresponding first few elements of the
dependent 1 L+1 X 2 , and the latter elements of the dependent
1 L+1 X 2 are weighted more heavily than the latter elements of
the independent X 1 . The parameter η 2 controls how quickly
the mixture moves from independent to dependent. The final
 Bayesian Distributed Lag Models
form for the prior on θ is then N (0, σ 2 Ω(η )), where Ω(η ) =
V(η 1 )W(η 2 )V(η 1 ) and η = (η 1 , η 2 ) .
Let θ̂ be the ML estimate of the unconstrained DL co-
efficients and let Σ be the sample covariance matrix. For a
normal linear DLagM, θ̂ is N (θ , Σ), so the posterior for θ
conditional on η and σ is
 −1
θ | θ̂ , η , σ 2 ∼ N 1/σ 2 Ω(η )−1 + Σ−1 Σ−1 θ̂ ,
 −1 
1/σ 2 Ω(η )−1 + Σ−1 . (1)
For a general linear DLagM, the posterior distribution for θ
may not be available in closed form, but it may be computed
through Gibbs sampling or other Markov chain Monte Carlo
methods (e.g., Carlin and Louis, 2000). We discuss such an
approach for our PM air pollution and mortality example, in
which the Y t are Poisson Ldistributed daily mortality counts,
log(E[yt | x1 , . . . , xt ]) = =0 θ xt− , and the likelihood for θ̂
is Poisson.
The influence of the prior distribution in estimating θ
depends on the values of hyperparameters σ 2 and η =
(η 1 , η 2 ) . The hyperparameter σ 2 , the prior variance of θ 0 ,
can be viewed as a tuning parameter determining the starting
point of the DL function. In practice there is little informa-
tion in the data to jointly estimate σ 2 and η . We therefore
assume σ 2 is ten times the estimated statistical variance of θ 0
so that even for relatively large values of η , the prior has little
to no influence on the first few DL coefficients. We examine
sensitivity of BDLagM estimates to choice of σ in Section 5.
Rather than setting values for η = (η 1 , η 2 ) and directly de-
termining the influence of the prior, we let η = (η 1 , η 2 ) have
a discrete uniform prior on N 1 × N 2 , where N 1 and N 2 are
finite sets of possible values for η 1 and η 2 . Then the poste-
rior distribution
 for θ can be defined as the weighted sum
p(θ | θ̂ ) = η p(θ | θ̂ , η )p(η | θ̂ ), where p denotes a general
probability density. Under the assumption that θ̂ ∼ N (θ , Σ),
the marginal posterior density of the hyperparameter η is
available in closed form. For a given η ∗ :
 
1 
| σ 2 Ω(η ∗ )Σ−1 + I| −1/2 exp − θ̂
2 σ2
p(η ∗ | θ̂ ) =  
. (2)
−1 −1/2 1 
| σ Ω(η )Σ
2
+ I| exp − θ̂
2 σ
η
Sufficiently large ranges for N 1 and N 2 insure that the
data drive the strength or weakness of the prior distribution
and therefore the eventual smoothness of the estimated DL
function.
3. Bayesian DLagMs and Penalized Splines
Following the well-established connection between nonpara-
metric smoothing and Bayesian modeling (e.g., Silverman,
1985), we illustrate the relationship between normal linear
BDLagMs and p-spline DLagMs. We show that estimating
the normal linear DL function under model (1) is analogous
to fitting a p-spline to DL coefficients with penalty derived
from our prior. An advantage of this connection is that our
method of putting a prior directly on the coefficients may be
viewed as a transparent means for eliciting p-spline penalties,
3
which are otherwise difficult to relate to biological or other
prior knowledge.
Let θ = U γ , where U is a spline basis matrix and γ
is a vector of spline coefficients. Let θ̂ be the ML esti-
mate of θ , and assume that θ̂ = U γ + ν , ν ∼ N (0, Σ), where
Σ is the estimated covariance matrix for θ̂ . Under a p-
spline approach, we estimate γ by minimizing the criterion
(θ̂ − U γ ) Σ−1 (θ̂ − U γ ) + λγ T D γ , where λ is a penalty pa-
rameter and D a positive semidefinite matrix (Eilers and
Marx, 1996; Ruppert et al., 2003).
To show the connection between minimizing this criterion
and estimating the BDLagM, (1), we reformulate the p-spline
in its Bayesian form θ̂ | γ ∼ N (U γ , Σ) and γ ∼ N (0, Γ),
where Γ is the prior covariance matrix of γ . Because θ =
U γ , the prior on γ translates to prior θ ∼ N (0, U ΓU  ). In
(1) we assume θ ∼ N (0, σ 2 Ω(η )), so we need Γ such that
U ΓU  = σ 2 Ω(η ), or Γ(η ) = R −1 Q  σ 2 Ω(η )Q R −1 where QR
is U’s qr-decomposition.
Under this formulation the log posterior for γ
is, up to a constant, − 12 (θ̂ − U γ) Σ−1 (θ̂ − U γ) −
1   
2
γ U (U Γ(η )W )−1 U γ , and maximizing the log poste-
rior for γ is equivalent to minimizing the above criterion with
λ = 1 and D = U  (U Γ(η ) W  )−1 U (Silverman, 1985; Green
and Silverman, 1994). For a given value of the hyperparame-
ter η , the estimated DL coefficients are given by the posterior
mean U (U  Σ−1 U + U  (U Γ(η )U  )−1 U −1 )−1 U  Σ−1 θ̂ , and the
equivalent degrees of freedom equal the trace of the smoother
matrix X (X TΣ−1 X + X T(X Γ(η )X T)−1 X −1 )X TΣ−1
(Ruppert et al., 2003).
Though a prior on DL coefficients may be translated to
a specific p-spline penalty, the spline approach requires that
the DL function follow a specific form, θ = U γ . For our air
pollution mortality example, we found that using a b-spline
basis with L + 1 degrees of freedom produced estimates of θ
identical to those from the BDLagM. In the following simula-
tion study, we compare BDLagMs to p-splines with penalties
unrelated to the prior.
 −1
1
Σ−1 − Σ−1 Σ−1 + Ω(η ∗ )−1 Σ−1 θ̂
  −1
−1 −1 −1 1 −1
Σ −Σ Σ + 2 Ω(η )−1 Σ θ̂
4. Simulation Study
We conducted a simulation study to compare BDLagMs with
four methods for estimating DL functions—unconstrained,
polynomial, p-splines with penalty parameter chosen by GCV,
and p-splines estimated with REML. We generated data un-
der 25 different sets of true DL coefficients, including examples
for which coefficients do not decrease to zero and smoothness
does not increase with lag. We categorize the DL functions
by four characteristics: (1) shape—decaying exponential (E),
step function (St), or gamma distribution (G); (2) latency—
0 or 2, the number of initial coefficients equal to zero; (3)
oscillation—as described by (−1) mod 2, to mimic mortality
displacement; and (4) maximum nonzero lag−7 or 14, the lag
4 Biometrics
by which the coefficients are less than 0.01. We also considered
a null DL function with all zero coefficients. All DL functions
included current day ( = 0). We set L = 14 as in the sub-
sequent air pollution mortality example. Except for the null
model, all the DL functions were normalized so the sum of
squares of the DL coefficients is 1. We refer to the nonnull
functions by [Shape]o ([latency], [max lag]), where the super-
script indicates oscillation.
Under each of the 25 scenarios, we generated 500 outcome
14
series y t from the model yt = δ =0 θ xt− + t where  t ∼
i.i.d. N(0,1), and δ is a constant to balance signal and noise.
For the exposure series x t we used mean centered PM 10 for
1996 from Chicago, Illinois because there were no missing ob-
servations and the autocorrelation is similar to what we ex-
perience when estimating the association between PM 10 and
mortality for Chicago for 1987–2000. For simplicity we take
the  t to be independent N (0, 1), noting that our simulations
still apply to situations in which the  t are autocorrelated be-
cause application of an appropriate linear filter will result in
a new DLagM with independent normal errors. We setδ =
0.25 to generate moderate evidence for a total effect,
in nonnull models (we empirically determined that δ = 0.25
generates y t such that the t-statistic for the ML estimate for

θ is approximately two). Similarly we set δ = 0.475 to
 
generate strong evidence for total effect (we empirically de-
termined that δ = 0.475 generates
 y t such that the t-statistic
for the ML estimate for 
θ  is approximately four). For
each simulated data set we compared the DL functions un-
der five methods: (1) unconstrained ML; (2) the proposed
Bayes’ method (Bayes) using the normal posterior as in (1);
(3) ML with a polynomial of degree four (Poly); (4) a pe-
nalized spline with penalty chosen by GCV (GCV); and (5)
a penalized spline estimated with REML (REML). We also
considered estimating the DL function using an AR-1 model.
With the exception of the null model and St0 (2, 14), the AR-1
model was not competitive, and was substantially worse when
the DL function oscillates then goes to zero.
Figure 1 shows the estimated DL functions (white) av-
eraged across the 500 simulations with the 95% confidence
bands (gray) for 24 of the true DL functions (black) (results
not pictured for null model). Results are reported for δ =
0.25. Visual inspection of this figure indicates that the BD-
LagM performs consistently well and estimates the true DL
function with narrower confidence bands than other methods.
To quantify the comparison, we summarize  the mean
squared errors of the estimated total effect ( θ ) and DL
coefficients at lags 0, 7, and 14 under the five estimation meth-
ods and for the 25 scenarios. Table 1 summarizes the results
for δ = 0.25. Results for δ = 0.475 are available in Web Ta-
ble 1. Mean squared errors are expressed as percentages of
the mean squared error of the corresponding unconstrained
ML estimates. Values smaller than 100 favor the proposed
estimation methods with respect to unconstrained ML.
When the DL function decreases to zero, BDLagM is 10 to
15% better at estimating the total effect than ML, whereas
Poly, GCV, and REML perform comparably to ML. Results
are similar for δ = 0.25 and δ = 0.475. The better performance
of the Bayesian method with respect its competitors is mainly
due to its greater flexibility in estimating the DL coefficients
at the longer lags. Bayes is consistently 20–30% better than
ML for lag 0; GCV and REML may be substantially better or
substantially worse. However, Bayes consistently outperforms
the others in estimating the lag 7 and the lag 14 coefficients
for scenarios in which the coefficients go to zero by lag 7 or 14.
When the BDLagM is misspecified and the DL coefficients do
not decrease smoothly to zero, performance of the BDLagM is
less predictable. Bayes may estimate the total effect only 5%
worse than ML (and Poly and REML), or nearly 15% better
(superior to Poly, GCV, REML).
Mortality counts are often modeled with Poisson log-linear
regression, so we also examine how our results extend to
the Poisson case. We simulated data from Y t ∼ Poisson(µ t ),
log(µ t ) = log(100) + Σ=14
=0 x t− θ  /100. The offset and division
by 100 were determined empirically to approximate Chicago
mortality levels in 1996. For each set of DL coefficients, we
generated 1000 mortality series. We estimated the posterior
distribution for θ two ways—using (1) (approximating θ̂ as
normal) or a Gibbs sampler. Web Table 2 compares the mean
squared errors of the total effects. The errors are comparable,
suggesting that the simulation results for normal outcomes
are not necessarily misleading for Poisson outcomes.
θ ,
5. Application to Particulate Matter Air Pollution
and Mortality
In this section, we apply BDLagMs to daily time series of
PM with aerodynamic diameter less than 10 microns (PM 10 )
and nonaccidental deaths for Chicago, Illinois for the period
1987–2000. The data were collected from publicly available
sources as part of the NMMAPS. NMMAPS contains daily
time series of age classified mortality, temperature, dew point,
and PM 10 for 109 U.S. cities from 1987 to 2000. We ana-
lyzed the time series for Chicago because it is the largest U.S.
city in NMMAPS with few missing PM 10 values. Additional
details regarding NMMAPS data assembly are available at
http://www.ihapss.jhsph.edu/ and are discussed in previ-
ous NMMAPS analyses (Samet, Zeger, Dominici, Curriero,
Dockery, Schwartz, and Zanobetti, 2000; Samet, Zeger, Do-
minici, Schwartz, and Dockery, 2000; Dominici et al., 2003).
Poisson log-linear regression is frequently used to estimate
the association between day-to-day variations in mortality
counts and day-to-day variations in ambient air pollution lev-
els. We accordingly assume that the mortality in Chicago on
day t, t = 1, . . . , 5114, is a Poisson random variable Y t with
expectation E[Yt ] = µ t . As above, we let θ = (θ 0 , . . . , θ L )
be the unknown DL coefficients we wish to estimate. We let
x t denote the PM 10 time series and for t > L we let x t de-
note the length L + 1 vector of lagged PM 10 values (x t , . . . ,
x t−L ) .
Multisite time series studies of single day exposure PM 10
and mortality have found strong evidence of an association
between PM 10 at lags l = 0, 1, and 2 and daily mortality
(e.g., Zmirou et al., 1988; Burnett, Cakmak, and Brook, 1998;
Katsouyanni et al., 2001; Dominici et al., 2003); single city
studies with DLagMs have similarly found the largest effects
in the first seven lags (e.g., Schwartz, 2000; Zanobetti et al.,
2003; Goodman et al., 2004). Though lags beyond two weeks
may have some influence on daily mortality (e.g., mortality
displacement), it is unlikely that lags beyond 2 weeks have
substantial influence on mortality compared to lags less than
2 weeks (Zanobetti et al., 2003). Models containing lags be-
yond 2 weeks are additionally difficult to estimate because
long-term averages of PM 10 have strong seasonal variation.
 Bayesian Distributed Lag Models 5

Figure 1. Mean estimated DL functions (white) and 95% posterior bands (gray) under five estimation methods—
unconstrained ML, the proposed Bayesian method (Bayes), ML with a polynomial of degree four (Poly), a penalized spline
with penalty chosen by GCV (GCV), and a penalized spline estimated with REML (REML). Outcome series were simulated
under moderately strong evidence for the sum of the DL coefficients (δ = 0.25).

We set L = 14 to capture the majority of short-term effects
of PM 10 on mortality without confounding estimation of DL
coefficients with seasonal trends in mortality.
When estimating air pollution health effects from time se-
ries studies it is important to account for potential time-
varying confounders such as weather, seasonality, and in-
fluenza epidemics (e.g., Schwartz, 1993; Samet et al., 1998;
Braga, Zanobetti, and Schwartz, 2000; Samoli et al., 2001;
Bell, Samet, and Dominici, 2004; Dominici, McDermott, and
Hastie, 2004; Peng, Dominici, and Louis, 2005; Welty and
Zeger, 2005). We let z t denote the vector of time-varying co-
variates to include in the model, and we specify z t as in pre-
vious NMMAPS analyses (Dominici et al., 2003). The exact
specification is documented in the associated R code, availa-
ble at http://www.ihapss.jhsph.edu/software/BayesDLM/.
Our goal is to estimate the DL coefficients θ as part of the
generalized linear model
log(µt ) = x t θ + z t β . (3)
The estimate for 1000 × θ  corresponds to the percentage
increase in daily mortality associated with a 10µg/m3 increase
14
in PM 10 at lag , and 1000 × =0 θ corresponds to the per-
centage increase in daily mortality associated with a 10µg/m3
increase in PM 10 at lags  = 0, . . . , 14.
Bayesian estimation of the generalized linear model in (3)
with our proposed prior for the DL coefficients θ requires two
6 Biometrics

Table 1
Mean squared errors of the estimates of the total effect and of the DL coefficients at lags 0, 7, and 14 obtained under four
estimation methods (Bayesian method (B), a polynomial with four degrees of freedom (P), a p-spline with penalty parameter
chosen by GCV (G), and a p-spline estimated with REML (R)) and for the 25 true DL functions. These results are reported
under the assumption of moderately strong evidence of a total effect (δ = 0.25). Mean squared errors are expressed as
percentages of the mean squared error of the corresponding ML estimates.

Total effect Lag 0 Lag 7 Lag 14

B P G R B P G R B P G R B P G R

E(0,7) 89 99 102 99 84 56 175 129 6 14 36 6 2 100 83 62

E(2,7) 91 99 100 99 78 47 59 77 9 11 31 16 2 135 94 102
E(0,14) 91 99 103 99 81 47 161 57 6 13 36 8 3 96 89 67
E(2,14) 95 99 99 99 78 70 56 62 8 11 22 15 6 108 95 98
o
E (0, 7) 89 99 100 99 81 58 119 167 6 22 42 10 1 129 92 78
Eo (2, 7) 89 99 100 99 77 43 70 76 7 16 47 12 2 141 96 89
Eo (0, 14) 89 99 100 99 80 48 74 162 15 50 37 26 2 134 96 70
Eo (2, 14) 88 99 100 99 74 44 81 49 11 50 58 18 3 124 102 83
St(0,7) 97 99 102 99 75 55 76 29 40 29 27 40 9 130 103 69
St(2,7) 99 99 98 99 74 88 40 49 50 38 23 38 10 126 86 75
St(0,14) 106 99 102 99 73 47 58 10 7 13 19 3 28 95 96 37
St(2,14) 105 99 96 99 72 59 29 24 7 13 25 6 30 95 76 61
Sto (0, 7) 87 100 100 99 82 67 68 113 98 206 41 187 4 96 99 50
Sto (2, 7) 87 100 100 100 73 61 72 24 46 179 51 220 5 97 102 37
o
St (0, 14) 86 99 100 99 81 52 65 84 72 183 70 135 180 355 99 248
Sto (2, 14) 86 99 99 99 73 43 65 15 33 133 31 142 188 316 93 339
G(0,7) 92 99 99 100 73 70 64 149 11 11 19 22 3 131 93 106
G(2,7) 92 100 99 100 75 187 55 94 16 28 27 33 4 96 86 84
G(0,14) 99 99 97 99 75 57 27 40 8 15 23 10 14 96 82 84
G(2,14) 99 99 100 99 75 89 25 71 18 18 27 11 20 143 93 71
o
G (0, 7) 88 100 100 99 71 73 86 63 7 27 60 9 2 134 106 42
Go (2, 7) 87 99 99 99 74 42 85 13 10 15 69 5 3 103 108 38
Go (0, 14) 87 99 100 99 76 50 80 41 63 180 59 109 3 100 96 40
Go (2, 14) 86 100 99 99 71 48 74 20 47 205 48 259 5 115 92 35
Null 89 99 96 99 74 47 21 10 5 13 24 3 1 95 83 37

extensions from the general approach outlined in Section 2.
First, the likelihood for (Y t | x t , z t ) is Poisson, so that θ̂ ,
the ML estimates of θ , will not be normal and the posterior
distribution for θ | θ̂ will not have a closed form expression.
Second, usual Bayesian estimation requires specifying a joint
prior for θ and β , an untenable approach given the size of the
nonpollutant covariate matrix and its potential relationship
with the pollutant covariate matrix.
We propose two approaches. The first is to fit (3) and
treat the ML estimates θ̂ as N (θ , Σ), where Σ is the sam-
ple covariance matrix. This approach ignores the uncertainty
introduced by estimating β and relies on the asymptotic
normality of the Poisson likelihood, but allows us to esti-
mate θ directly using its closed form posterior (1). The sec-
ond approach is to fit the Poisson log-linear model using
a Gibbs sampler; details and code are available at http://
www.ihapss.jhsph.edu/software/BayesDLM/.
For both computational methods, we set the hyperprior on
η = (η 1 , η 2 ) to be a discrete uniform distribution over N 1 ×
N 2 , where N 1 is a length 10 sequence ranging from −0.35
to −0.05 in equal intervals, and N 2 is a length 10 sequence
ranging from −0.37 to 0 in equal intervals. We selected the in-
terval for N 1 so that the ratio of the prior standard deviation
of θ 0 to θ L is bounded between 2 and 100. We selected the
values for N 2 so that the prior correlation of θ L−1 and θ L is
bounded approximately by 0 and 0.99. We also set σ = 0.004,
slightly larger than the square root of ten times the estimated
variance in the ML estimate of θ 0 . The sensitivity of the esti-
mated BDLagMs to choices of σ and N 1 × N 2 is considered
below. We ran the Gibbs sampler for K = 5000 iterations, dis-
carding the first 1000 as burn-in. Diagnostic checks suggested
that the algorithm converged.
Figure 2 shows the posterior mean and the 95% posterior
region of the DL function for the association between PM 10
 Bayesian Distributed Lag Models 7

Figure 2. Posterior mean (white) of the DL function for the effect of PM 10 on mortality for Chicago, Illinois from 1987 to
2000, using the last 4000 of 5000 iterations of the Gibbs sampler. The gray shaded region denotes the 95% posterior region.
Black dots indicate ML estimates for the unconstrained DL coefficients.

and mortality in Chicago from 1987 to 2000. The black dots
indicate the unconstrained ML estimates of the DL coeffi-
cients. The strongest association between PM and mortality
occurs at lag 3: a 10µg/m3 increase in PM 10 at lag 3 is associ-
ated with a 0.17% increase in mortality (95% posterior inter-
val [PI] 0.01%, 0.34%), all other lagged PM 10 levels remaining
constant. The drop in relative risk from lag 3 to lag 5 suggests
the possibility of mortality displacement. We estimate a to-
tal effect of −0.24% (95% PI −0.73%, 0.23%). The estimated
total effect using unconstrained ML, −0.19%, is similar, but
has a wider 95% confidence interval (−0.86%, 0.48%). The
joint posterior distribution of η = (η 1 , η 2 ) (see Web Figure
1) favored models for which Var(θ  ) → 0 quickly and Cor(θ  ,
θ +1 ) → 1 moderately or quickly.
Figure 3 compares posterior distributions of DL coefficients
from the Gibbs sampler (black) and the normal approxima-
tion (gray). The estimates from the two methods differ for
more moderate lags but are similar for early and later lags and
for the overall sum of DL coefficients. This pattern of agree-
ment and discrepancy is not surprising, given that we expect
the normal approximation and the true posterior distribution
to be most similar where the prior is weakest and the data
drive estimation (early lags) and where the prior is strongest
and drives estimation (later lags). The normal approxima-
tion was computationally faster than the Gibbs sampler (on
an AMD Opteron 848 system with a 2.2 GHz processor, 8.6
seconds versus 15.5 hours for 5000 iterations).
We examined the sensitivity of the BDLagM estimates to
the specification of the prior on η and the selection of the
value for σ (Web Figure 2). The value for σ 2 was initially set
to 10 times the estimated variance of θ 0 . Larger values of σ re-
sult in BDLagMs that more closely followed the unconstrained
ML estimates at longer lags. Smaller values of σ resulted in
BDLagMs with latter DL coefficients shrunk to zero. For σ =
0.04, 0.004, 0.0004, the initial DL coefficient estimates were
indistinguishable. The original discrete uniform prior on η 1
was set so that the ratio of the prior standard deviation of θ 0
to θ L ranged approximately from 2 to 100. We considered two
new priors for η 1 so that the ratio ranged from approximately
2 to 50 (more restrictive) or from 2 to 200 (less restrictive). We
did not consider alternate priors on η 2 because the prior was
already constructed to be as broad as possible without cre-
ating numerical instability. The BDLagMs estimated across
different prior distributions for η and σ = 0.004 were remark-
ably similar. We concluded that the estimated BDLagM is
not driven strongly by the range of values for η .
6. Discussion
We introduce a Bayesian approach to estimate DL functions
in time series models of air pollution and mortality. This for-
mulation uses prior knowledge about the shape of the DL
function, and allows the degree of smoothness of the DL
function to be estimated from the data. We illustrate in a
simulation study that when prior assumptions are valid, BD-
LagMs estimate DL coefficients with smaller mean squared
errors than three common methods—polynomial, spline, and
unconstrained DLagMs.
8 Biometrics

Figure 3. Comparison of estimation methods for DL coefficients of the effect of PM 10 on mortality for Chicago, Illinois from
1987 to 2000 by estimation method. Distributions of DL coefficient estimates, by lag, and sum of DL coefficients (all in units
of 10−4 ) are shown for (i) the DL coefficient vector simulated from the normal approximate posterior distribution (gray) and
(ii) the estimates of DL coefficients from last 4000 iterations of the Gibbs sampler (black).

We also show that our approach relates to using penalized
splines to estimate DL functions. Specifically, fitting a penal-
ized spline DLagM with a specific penalty matrix is analogous
to using a BDLagM with a normal prior on the DL coefficients.
An advantage of using the Bayesian approach is the simplicity
of formulating a prior distribution on DL coefficients rather
than specifying a penalty matrix.
Using the proposed BDLagM we estimated the association
between lagged exposures of PM 10 and mortality for Chicago,
Illinois from 1987 to 2000. We found that the largest effect of
PM 10 on mortality occurs at lag 3 and that the total effect is
equal to −0.21% (95% PI −0.86%, 0.41%). The shape of the
DL function is consistent with mortality displacement.
For the Chicago data we found that the BDLagM esti-
mated using the normal approximation to the likelihood (with
a posterior distribution for θ available in closed form) and
the Poisson likelihood (with a Gibbs sampler) yielded simi-
lar estimates for the total effect and for early and later DL
coefficients. The relatively large number of daily deaths in
Chicago (on average, 116) as well as the length of the time
series may account for the agreement between the two meth-
ods. For applications with outcome distributions that are not
approximately normal, we anticipate less agreement between
the two estimation methods and that the normal approximate
posterior would be a less efficient proposal distribution.
The BDLagM formulated for a single city time series study
may be naturally extended to a multicity framework. Multi-
city studies of mortality and air pollution use hierarchical
models to pool individual city relative risks across multiple
cities or counties, and have provided strong evidence for the
association between air pollution and mortality (Zmirou et al.,
1988; Burnett et al., 1998; Schwartz, 2000, Katsouyanni et
al., 2001; Samoli et al., 2001; Zanobetti et al., 2002, 2003;
Dominici et al., 2003). The hierarchical models used to date
have estimated risk for single lag PM exposures or the total
effect, which may not fully describe the relationship between
short-term health risk and air pollution exposure. Estimating
our BDLagM for multiple cities in a hierarchical model of an
overall DL function between air pollution and mortality would
provide additional understanding of the relationship between
air pollution and health (Peng, Dominici, and Welty, 2007).
A challenge to estimating our BDLagMs for multiple cities
is missing data. For many U.S. cities, PM air pollution is
measured 1 in every 6 days. Before estimating the outlined
 Bayesian Distributed Lag Models
BDLagMs for multicity studies, it will be necessary to de-
velop a version that estimates the DL coefficients in the pres-
ence of missing data. Accounting for missingness in the ex-
posure series would expand the applicability of the proposed
BDLagMs.
Given the equivalence between estimating DL functions us-
ing a penalized spline and putting a prior directly on the DL
coefficients, our Bayesian method may be viewed as a means
for eliciting a penalty matrix. P-spline penalties can be in-
terpreted as the size of jumps of a smooth’s third or higher
derivatives, which may be difficult to relate to biological or
other prior knowledge. Our method may be viewed as a trans-
parent or intuitive means for eliciting penalties that are con-
sistent with prior knowledge of the objective function. Our
approach is not limited to functions that increase in smooth-
ness as they approach zero; it could also be applied, for in-
stance, to monotonic functions. However, given the necessity
of choosing a value for σ 2 = Var(θ 0 ), it could be imprudent to
use our approach to estimate DL functions about which there
is no prior knowledge about the range of θ 0 .
7. Supplementary Materials
Web Tables and Figures referenced in Sections 4 and 5 are
available under the Paper Information link on the Biometrics
website at http://www.biometrics.tibs.org.
Acknowledgements
Funding for the authors was provided by NIEHS RO1 grant
(ES012054-01), and by NIEHS Center in Urban Environmen-
tal Health (P30 ES 03819).
References
Almon, S. (1965). The distributed lag between capital appro-
priations and expenditures. Econometrica 33, 178–196.
Bell, M. L., Samet, J. M., and Dominici, F. (2004). Time-
series studies of particulate matter. Annual Review of
Public Health 25, 247–280.
Bell, M. L., McDermott, A., Zeger, S. L., Samet, J. M., and
Dominici, F. (2004). Ozone and short-term mortality in
95 US urban communities, 1987–2000. Journal of the
American Medical Association 19, 2372–2378.
Braga, F., Zanobetti, A., and Schwartz, J. (2000). Do respi-
ratory epidemics confound the association between air
pollution and daily deaths? European Respiratory Jour-
nal 16, 723–728.
Braga, F., Luis, A., Zanobetti, A., and Schwartz, J. (2001).
The time course of weather-related deaths. Epidemiology
12, 662–667.
Burnett, R., Cakmak, S., and Brook, J. (1998). The effect of
the urban ambient air pollution mix on daily mortality
rates in 11 Canadian cities. Canadian Journal of Public
Health 89, 152–156.
Carlin, B. P. and Louis, T. A. (2000). Bayes and Empirical
Bayes Methods for Data Analysis. Boca Raton, Florida:
Chapman and Hall.
Corradi, C. (1977). Smooth distributed lag estimators and
smoothing spline functions in Hilbert spaces. Journal of
Econometrics 5, 211–220.
9
Dominici, F., McDermott, A., Daniels, M. J., Zeger, S. L.,
and Samet, J. M. (2003). Revised Analysis of the Na-
tional Morbidity Mortality Air Pollution Study: Part II.
Cambridge, Massachusetts: The Health Effects Institute.
Dominici, F., McDermott, A., and Hastie, T. (2004). Im-
proved semi-parametric time series models of air pollu-
tion and mortality. Journal of the American Statistical
Association 468, 938–948.
Eilers, P. and Marx, B. (1996). Flexible smoothing with
b-splines and penalties. Statistical Science 1, 89–121.
Fahrmeir, L. and Knorr-Held, L. (1997). Dynamic discrete-
time duration models: Estimation via Markov chain
Monte Carlo. Sociological Methodology 27, 417–452.
Goodman, P. G., Dockery, D. W., and Clancy, L. (2004).
Cause-specific mortality and the extended effects of par-
ticulate pollution and temperature exposure. Environ-
mental Health Perspectives 112, 179–185.
Green, P. J. and Silverman, B. W. (1994). Nonparametric
Regression and Generalized Linear Models. Boca Raton,
Florida: Chapman and Hall.
Hastie, T. and Tibshirani, R. (1993). Varying-coefficient mod-
els. Journal of the Royal Statistical Society, Series B 4,
757–796.
Katsouyanni, K., Toulomi, G., Samoli, E., Gryparis, A.,
Le Tertre, A., Monopolis, Y., Rossi, G., Zmirou, D.,
Ballester, F., Boumghar, A., Anderson, H. R., Woj-
tyniak, B., Paldy, A., Braunstein, R., Pekkanen, J.,
Schindler, C., and Schwartz, J. (2001). Confounding and
effect modification in the short-term effects of ambi-
ent particles on total mortality: Results from 29 Euro-
pean cities within the APHEA2 project. Epidemiology
12, 521–531.
Kim, H., Kim, Y., and Hong, Y. (2003). The lag-effect pat-
tern in the relationship of particulate air pollution to
daily mortality in Seoul, Korea. International Journal of
Biometeorology 48, 25–30.
Leamer, E. E. (1972). A class of informative priors and dis-
tributed lag analysis. Econometrica 40, 1059–1081.
Manda, S. and Meyer, R. (2005). Age at first marriage in
Malawi: A Bayesian multilevel analysis using a discrete
time-to-event model. Journal of the Royal Statistical So-
ciety, Series A 168, 439–455.
Peng, R. D., Dominici, F., and Louis, T. A. (2005). Model
choice in time series studies of air pollution and mortal-
ity. Journal of the Royal Statistical Society, Series A 169,
179–203.
Peng, R. D., Dominici, F., and Welty, L. J. (2007). A Bayesian
hierarchical model for constrained distributed lag func-
tions: Estimating the time course of hospitilization asso-
ciated with air pollution exposure. Technical Report 128,
Department of Biostatistics, Johns Hopkins University,
Baltimore, Maryland.
Pope, C. A. and Schwartz, J. (1996). Time series for the
analysis of pulmonary health data. American Journal
of Respiratory and Critical Care Medicine 154, S229–
S233.
Pope, C. A. R., Dockery, D. W., Spengler, J. D., and
Raizenne, M. E. (1991). Respiratory health and pm10
pollution. A daily time series analysis. American Review
of Respiratory Diseases 144, 668–674.
10 Biometrics
Ravines, R. R., Schmidt, A. M., and Migon, H. S. (2006). Re-
visiting distributed lag models through a Bayesian per-
spective. Applied Stochastic Models in Business and In-
dustry 22, 193–210.
Roberts, S. (2005). An investigation of distributed lag models
in the context of air pollution and mortality time series
analysis. Journal of the Air and Waste Management As-
sociation 55, 273–282.
Ruppert, D., Wand, M. P., and Carroll, R. J. (2003).
Semiparametric Regression. Cambridge, U.K.: Cambridge
University Press.
Samet, J., Zeger, S., Kelsall, J., Xu, J., and Kalkstein, L.
(1998). Does weather confound or modify the association
of particulate air pollution with mortality. Environmental
Research 77, 9–19.
Samet, J. M., Zeger, S. L., Dominici, F., Curriero, F., Dock-
ery, D. W., Schwartz, J., and Zanobetti, A. (2000). The
National Morbidity Mortality Air Pollution Study: Part II.
Cambridge, Massachusetts: The Health Effects Institute.
Samet, J. M., Zeger, S. L., Dominici, F., Schwartz, J., and
Dockery, D. W. (2000). The National Morbidity Mortality
Air Pollution Study: Part I. Cambridge, Massachusetts:
The Health Effects Institute.
Samoli, E., Schwartz, J., Wojtyniak, B., et al. (2001). Inves-
tigating regional differences in short-term effects of air
pollution on daily mortality in the APHEA project: A
sensitivity analysis for controlling long-term trends and
seasonality. Environmental Health Perspectives 109, 349–
353.
Schiller, R. J. (1973). A distributed lag estimator derived from
smoothness priors. Econometrica 41, 775–788.
Schimmel, B. and Murawsky, T. (1978). The relation of air
pollution to mortality. Journal of Occupational Medicine
18, 316–333.
Schwartz, J. (1993). Methodological issues in studies of air
pollution and daily counts of deaths or hospital admis-
sions. American Journal of Epidemiology 137, 1136–1147.
Schwartz, J. (2000). The distributed lag between air pollution
and daily deaths. Epidemiology 11, 320–326.
Silverman, B. W. (1985). Some aspects of the spline smooth-
ing approach to non-parametric regression curve fitting.
Journal of the Royal Statistical Society, Series B 47, 1–
52.
Welty, L. J. and Zeger, S. L. (2005). Are the acute effects
of particulate matter on mortality in the National Mor-
bidity, Mortality, and Air Pollution study the result of
inadequate control for weather and season? A sensitivity
analysis using flexible distributed lag models. American
Journal of Epidemiology 162, 80–88.
Zanobetti, A., Wand, M. P., Schwartz, J., et al. (2000). Gener-
alized additive distributed lag models: Quantifying mor-
tality displacement. Biostatistics 1, 279–292.
Zanobetti, A., Schwartz, J., Samoli, E., Gryparis, A.,
Touloumi, G., Atkinson, R., Le Tertre, A., Bobros,
J., Celko, M., Goren, A., Forsberg, B., Michelozzi, P.,
Rabczenko, D., Aranguez Ruiz, E., and Katsouyanni, K.
(2002). The temporal pattern of mortality responses to
air pollution: A multicity assessment of mortality dis-
placement. Epidemiology 13, 87–93.
Zanobetti, A., Schwartz, J., Samoli, E., Gryparis, A.,
Touloumi, G., Peacock, J., Anderson, R. H., LeTertre,
A., Bobros, J., Celko, M., Goren, A., Forsberg, B., Mich-
elozzi, P., Rabczenko, D., Hoyos, S. P., Wichmann, H.
E., and Katsouyanni K. (2003). The temporal pattern
of respiratory and heart disease mortality in response
to air pollution. Environmental Health Perspectives 111,
1188–1193.
Zmirou, D., Schwartz, J., Saez, M., Zanobetti, A., Wojty-
niak, B., Touloumi, G., Spix, C., Ponce de León, A., Le
Moullec, Y., Bacharova, L., Schouten, J., Pönkä, A., and
Katsouyanni, K. (1988). Time-series analysis of air pol-
lution and cause-specific mortality. Epidemiology 9, 495–
503.
Received December 2005. Revised January 2008.
Accepted January 2008.